JULY 1967
SYNTHESIS OF (+)-a-
18-in. spinning-band column, collecting the fraction (9.5 g) of bp 140-170'. Vpc analysis (on a column of DC 710 supported on Chromosorb W, 20 mesh, at 165" with a helium flow rate of 75 ml/min) showed two peaks with retention times of 29 and 32 min, in the ratio 2 : 5 . Samples of both substances were collected by repeated injection on the same column. The faster moving component was identified as methyl-trans2-methylcyclopentanecarboxylate,[a]2 6 ~- 57.5" (c 0.2, carbon tetrachloride). I t was converted to the amide by bubbling dry ammonia through a solution of the ester (120 mg) in 5 ml of methanol for 30 min, then keeping the solution sealed a t room temperature for 3 days. The solvent was removed in a stream of dry nitrogen and the residue was crystallized from benzenepetroleum ether (bp 30-60'). The amide (XVII) had mp 149.0149.5", [ a I z 6 ~ -40.2' ( c 0.0076, water).3E Hooper, et al.,6 reported mp 152-153.5", [ a ] 2 0-13.7' ~ (c 1,water), for the amide prepared from dihydrosarkomycin. The infrared and nmr (36) This rotation was measured on a Cary spectropolarimeter; we thank
Mr. Grant Krow for the measurement. (37) (a) W. 0.Ney, W. W. Crouch, C. E. Rannefeld, and H. L. Lochte, J. Am. Chem. Soc., 66, 770 (1943);(b) H.Pines and N. E. Hoffman, ihid., 76, 4417 (1954);(c) M.Julia and F. Le Goffic, BuEE. Soc. Chim. France, 1550 (1965). We are grateful to Professor M. Julia and Mr. M. Maumy for sending us a sample of racemic trans-2-methylcyclopentanecarboxylicacid.
AND
(+)-~-EUDESMOLS
2335
spectra of XVII were identical with those of an authentic sample of the racemic amide," mp 149-150". The second substance eluted from the vpc column was an isomeric ester [mol wt 142 (mass spectrum), [(Y]%D -6.8" (c 0.2, carbon tetrachloride)] which was converted in the same manner to the amide [mp 148.5-149.5", [ ( Y ] ~ ~-32.2' D ( c 0.01, water)] .*6 This is tentatively regarded as one of the stereoisomers of 3methylcyclopentanecarboxamide; the racemic amide (configuration unknown) melted*7*a t 147-148'.
Registry No.-( -)-I, 13012-37-8; (+)-111, 1301238-9; (+)-I11 2,4-dinitrophenylhydrazone, 13012-39-0; V, 13052-48-7; V p-nitrobenzoate, 13012-40-3; (+)-X, 7712-68-7; (-)-XI, 13012-42-5; (-)-XII, 13012-43-6; (-)-XIII, 13012-44-7 ; (-)-XIIT. bis-p-nitrobenzoate, 13012-45-8; (-)-XIV, 13012-46-9; (-)-XV, 1301247-0; (-)-XV bis-p-nitrobenzoate, 13012-48-1; (-)XVII, 13012-49-2; methyl trans-2-methylcyclopentanecarboxylate, 13012-50-5; 3-methylcyclopentanecarboxamide, 13012-51-6. Acknowledgment.-The authors express their thanks to Mr. Norman Gilman for obtaining the nmr spectra.
Synthetic Experiments in the Eudalene Group of Bicyclic Sesquiterpenes. 1II.l Total Synthesis of (+)-a- and (+)-P-Eudesmols D. c. H U M B E R , 2 v 3A. R. P I N D E R , I
AND
R. A.
WILLIAMSzp5
Departments of Chemistry, University College, Cardif, Wales, and Clemson University, Clemson, South Carolina Received January 27, 1967 The total syntheses of (+ >a-eudesmol and (+>@-eudesmolare described.
The natural product known as eudesmol has been recognized for many years to be a mixture of two and sometimes three isomeric alcohols, depending upon the source of the materiaL6 These alcohols are the double-bond isomers a-,p-, and y-eudesmols, proved to have structures 1, 2, and 3, respectively. The
2 represents (+)-p-eudesmol; a recent stereoselective total synthesis of (A)-p-eudesmol has provided confirmation.1° The structure of a-eudesmol (1) is based on its chemical transformations.6 I n this paper we describe stereoselective total syntheses of (+)-aeudesmol and (+)-@-eudesmol, which confirm structures and absolute configurations 1 and 2, respectively, for these alcohols."
Synthesis of (+)-p-Eudesmol 15
H
HISoH 1
2
3
constitution 3 assigned to y-eudesmol is based on its chemical transformations,' and it has been confirmed by total synthesis8 that 3 represents the structure and absolute configuration of (+)-y-eudesmol. The structure 2 for the 13 isomer is also based on its chemical behavior: and stereochemical studies9 have shown that (1) Part 11: D. C. Humber and A. R. Pinder, J . Org. Chem., Si, 4188 (1966). (2) Abstracted in part from theses presented by D. C. H. and R. A. W. in partial fulfilment of the requirements for the degree of Doctor of Philosophy in the University of Wales. (3) Glaxo Research Ltd., Greenford, Middlesex, England. (4) To whom enquiries should be addressed a t Clemson University. (5) Department of Chemistry, North Staffordshire College of Technology, Stoke-on-Trent, England. (6) For a review of eudesmol chemistry, see D. H. R. Barton, in "Chemistry of Carbon Compounds," Vol. IIB, E. H. Rodd, Ed., Elsevier Publishing Co., Amsterdam, 1953,p 664. (7) (a) F. J. McQuillin and J. D. Parrack, J . Chem. Soc., 2973 (1956); (b) R. B. Bates and E. K. Hendrickson, Chem. I n d . (London), 1759 (1962). (8) A. R.Pinder and R. A. Williams, J . Chem. Soc., 2773 (1963). (9) (a) A. J . Birch and K. M . C. Mostyn, Australtan J . Chem., 7, 301 (1954); (b) B. Riniker. J. Kalvoda, D. Arigoni, A. Farst, 0. Jeger, A. M. Gold, and R. B. Woodward, J . Am. Chem. Soc., 76, 313 (1954).
The starting point in our first approach to (+)-peudesmol was (+)-15-nor-a-~yperone~~ (4) obtainable by a Robinson-Mannich annelation reaction between (+)-dihydrocarvone and 1-diethylaminobutan-3-one methi0dide.l Compound 4 was a minor product of the annelation reaction; it was isolated through its semicarbazone by a rather tedious procedure, and in very poor yield. Nevertheless, sufficient material was obtained to make a synthetic sequence starting from it possible. The absolute configuration of 4 was settled by the observation that its ORD curve is almost superimposable on that of (+)-a-cyperone, of proven absolute configuration (5).13 Treatment of 4 (10) J. A. Marshall and M. T. Pike, Tetrahedron Letters, 3107 (1965); J. A. Marshall, M. T. Pike, and R. D. Carroll, J. Org. Chem., 81, 2933 (1966). (11) For preliminary accounts, see (a) R. P. Houghton, D. C. Humber, and A. R . Pinder, Tetrahedron Letters, 353 (1966); (b) D. C. Humber and A. R. Pinder, ibid., 4985 (1966). (12) We are using the numbering for the eudesmane ring system proposed by W. Cocker and T. B. H. McMurry, Tetrahedron, 8 , 181 (1960),as depicted in formula l. I n our earlier paper' we have described compound 4 as (+)-4-nor-u-cyperone; we now feel i t is better designated (+)-15nor-a-cyperone. (13) R. Howe and F. J. McQuillin, J . Chem. Soc., 2423 (1955).
HUMBER, PINDER, AND WILLIAMS
2336
a.
5
4
0
H
H
6
q
oH
H
7
with perbenzoic acid afforded the 1l112-epoxide6, which was reduced by lithium aluminum hydride to the diol 3-hydroxy-l&nor--peudesmol (7) as a mixture of Ca epimerides. Oxidation with activated manganese dioxideI4 gave (+)-15-norcarissone 8, which was converted to its ethylene dithioketal 9 on treatment with ethanedithiol and boron trifluoride etherate. Attempts to desulfurize the dithioketal 9 to 15-nor-yeudesmol (10) were, at this juncture, unsuccessful, the product
VOL.32
settled by a knowledge of the steric course1*of the annelation reaction, and by the fact that the ORD curve’@of the ketol 12 bore a mirror-image relationship to that of 14, described earlier.’ Further, the ORD curve of 12 is quite unlike that of coprostan-3-one1 in agreement with the expectation that 12 has nonsteroidal cis-fused rings, as in 15 (strong positive Cotton effect).20 The observed chemical shift (1.18 ppm) of the angular methyl group in 12 is in agreement with this assignment of conformation. 18,21 Accompanying the ketol 12 as minor products were the aldol-type compound’eb 16, and the octalones 17 and 18, 16 being separable from the ketone mixture by
OH
15
HO%‘ 8
9
cQ.f H
0
17
%OH 10
containing appreciable amounts of the corresponding fully saturated alcohol in each experiment. Satisfactory separation of these did not prove practicable. However, satisfactory conditions for this desulfurization were eventually discovered (see below). Attention was next turned to the possible use of (- )-dihydrocarvone, of established stereochemistry16 11, as starting point. A Robinson-Mannich reaction between the ketone and nascent methyl vinyl ketone (from 4-diethylaminobutan-3-one methiodide and methylsulfinyl carbanion16 in dimethyl sulfoxide) afforded the cis-keto1 12. The possibility that the product
H*Qy 0
H
11
13
oa3f OH H 12
14
might have the bridged structure” 13 was ruled out by the fact that its nmr spectrum showed only two singlet methyl proton signals, at 1.18 and 1.68 ppm (TMS = 0). The cis stereochemistry at the ring junction was (14) 0.Mancera, G. Rosenkranr, and F . Sondheimer, J . Chem. Soc., 2189 (1953). (15) A. J. Birch, Ann. Rept. Chem. Soc., 47, 192 (l950),and references cited therein. (16) (a) E. J. Corey and M. Chaykovsky, J . A m . Chem. Soc., 84, 866 (1962); (b) ibid., 87, 1345 (1965); (c) G. G. Price and M. C. Whiting, Chem. I n d . (London), 775 (1963). (17) Cf. W. S. Johnson, J. J. Korst, R. A. Clement, and J. Dutta, J . A m . Chem. Soe., 88,614(1960).
16
H 1’8
chromatography. The aldol is formulated as 16 because its physical and chemical properties are identical (except for its specific rotation, which is opposite in sign) with those of the enantiomorph described earlier.’ The ketones 17 and 18 were present in the ratio 1:2; a partial separation, leading t o pure 18, was possible through the mixed semicarbazones, but this compound was more conveniently obtained by acid-catalyzed dehydration of ketol 12. Ozonolysis of octalone 18 afforded the diketone 19, together with the epoxide 20, the products being sepac rable by preparative thin layer chromatography. The 4cu,5a configuration 20 for the epoxide is based on its correspondence in physical and chemical properties (with equal, but opposite, specific rotation) with the enantiomorphous product.’ The formation of 20 could be suppressed by effecting the ozonolysis in the presence of pyridine.22 Similarly, the assignment of the 7a configuration for the acetyl group in 19 is justified by the correspondence in spectral and optical rotatory dispersion properties between 19 and its enantiomer, described in our earlier paper.‘ Treatment of 19 with acid, as anticipated, induced epimerization at position 7, with formation of the 7@-acetyloctalone21. The driving force behind this facile inversion is presumably the fact that in 18, and therefore in its ozonoly(18) F. J. McQuillin, J . Chem. Soc., 528 (1955); T. A. Spencer, K. K. Schmiegel, and K. L. Williamson, J . A m . Chem. Soc., 86, 3785 (1963); J. A. Marshall and W. I. Fanta, J . Ore. Chem., 29, 2501 (1964); T . A. Spencer, H. 8. Neel, D . C. Ward, and K. L. Williamson, ibid., 31, 434 (1966); K. L. Williamson, L. R. Sloan, T . Howell, and T. A. Spencer, ibid., 91, 436 (1966). (19) We are greatly indebted to Professor W. Klyne, Dr. P. M. Scopes, and Dr. 8. Wallis (Westfield College, University of London) for ORD measurements and discussions. (20) Cf.R. Howe and F. J. McQuillin, J . Chem. Soc., 1194 (1958); T. G. Halssll, D. W. Theobald, and I(.B. Walshaw, ibid., 1029 (1964); C. Djerassi. “Optical Rotatory Dispersion,” McGraw-Hill Book Co., Inc., New York, N.Y..1960,p 185. (21) W. G. Dauben, R. M. Coates, N. D. Vietmeyer, L. J. Durham, and C. Djerami, Ezperientia, 21,565 (1965). (22) Cf.a.Slomp and J. L. Johnson. J . A m . Chem. Soc., 80, 915 (1958).
SYNTHESIS OF (+)-a-AND (+)-@-EUDESMOLS
JULY1967
sis product 19, ring B is a boat, to allow the C7 substituent to be equatorial. The ORD curve of 18 is very similar to that of epi-a-cyperone, 22, and of 8-isotestosterone; in both these compounds there are convincing reasons for believing that ring B is severely distorted so as to be roughly boat shaped.z3 Inversion a t C, in 19 permits ring B to assume the thermodynamically more stable chair conformation, with attendant equatorial 76-acetyl group, as in 2 1.
19
20
2337
aluminaz7was homogenized to the trans-1-decalone 27. The latter compound proved to be identical in every respect with the crystalline ketone obtained by the ozonolysis of commercial eudesmol.gaJ8 Ketone 27 had an ORD curve closely similar to that of cholestan-4onelz9 confirming the assigned stereochemistry. Finally, a Wittig reaction between 27 and methylenetriphenylphosphorane gave (+)-@-eudesmol (Z), purified by preparative thin layer chromatography and vacuum sublimation, and identical in all respects with natural (+)-@-eudesmol, separated from commercial eudesmol. 30 This synthesis is also a synthesis of cryptomeridi013~ and ne~intermedeol,~~ both of which have been synthesized from (+)-@-eudesmol. Synthesis of (+)-a-Eudesmol
0
o@-f
HO
22
21
The diketone 21 was also obtainable, in better yield, by ozonolysis of the keto1 12, to the diketo alcohol 23, followed by acid dehydration, with concomitant epimerization. The ORD curve of 23 resembled closely that of 12, indicating that the ozonolysis product was similar in conformation to 12.
The starting point in this synthesis was (+)-carissone (28), the structure33and absolute configurations*34 of which are firmly established. Reduction with lithium in liquid ammonia35 afforded a product consisting in the main of the expected trans-dihydrocarissone (29) accompanied by a small amount of the corresponding diol 30. The crude product was treated
29
30
N
I
NHTs
Condensation of diketone 21 with ethanedithiol (1 mole) resulted mainly in a selective attack2*on the conjugated carbonyl group, with formation of 24, accompanied by a small amount of the bis(ethy1ene dithioketal) 25. The two were separated by fractional crystallization, and 24, identified by its infrared spectrum and analysis, was treated with methylmagnesium iodide, to give the alcohol 9. Desulfurization of 9 with sodium and liquid ammoniaz5 furnished 15-nory-eudesmol (10) ; the use of a partially poisoned Raney nickel catalyst achieved substantially the same result. The ORD curve of this product was almost identical with that of y-eudesmols (3). Hydroboration of 15nor- y-eudesmol gave, after oxidation with hydrogen peroxide, a mixture of epimeric alcohols 26; this, on oxidation with Jones reagentz5 afforded a mixture of epimeric 1-decalones, which on exposure to basic
26
27
(23) C. Djerassi, R. Riniker, and B. Riniker, J . Am. Chem. Soc., 78, 6362 (1956). (24) Cf.J. W.Ralls and B. Riegel, ibid., 76, 4479 (1954). (25) Cf. R. E. Ireland, T. I. Wrigley, and W. G. Young, ibid., 80, 4604 (1958). (26) K. Bowden, I. M. Heilbron, E. R. H. Jones, and B. C. L. Weedon, J . Chem. Soc., 39 (1946).
31
with Jones reagent,z6 to give a homogeneous ketonic product 29, in which we assign the more stable CY configuration (equatorial) to the Cd-methyl group. The dihydro ketone 29 was subjected to a Bamford-Stevens reaction,36 being converted into its tosylhydrazone (31) which was heated with sodium ethylene glycolate. The product, after purification by sublimation, preparative thin layer chromatography, and glpc, proved to be identical in all respects with an authentic sample of (+)-a-eudesmol ( l ) , separated from commercial eudesmol essentially as described by Seikel and R0we,~7 and also by preparative thin layer chromatography. (27) Cf.H. B. Henbest and T. I. Wrigley, ibid., 4596 (1957). (28) L. Ruzicka, A. H. Wind, and D . R. Koolhaas, H e h . Chim. Acta, 14, 1132 (1931);A. J. Birch and K. M. C . Mostyns have also described the ketone, but wereunable to obtainit crystalline. (29) C. Djerassi, W. Closson, and A. E. Lippman, J . Am. Chem. Sac., 78, 3163 (1956); C. Djerassi, Bull. SOC.Chim. France, 741 (1957). The amplitude (-56) is of the order (-60) expected for a trans-1-decalone of this type: W. Klyne, Bzpcrientia, I O , 349 (1964). (30) (a) K. R. Varma and S. C. Bhattacharyya, Tetrahedron 10, 2927 (1964); we are indebted to Dr. Bhattacharyya, Poona, for an authentic specimen of (+)-@-eudesmol. (b) See also ref 7a (mp 76"). (31) M . Sumimoto, H. Ito, H. Hirai, and K. Wads, Chem. Ind., (London), 780 (1963). (32) V. B. Zalkow, A. M. Shaligram, and L. H. Zalkow. ibid., 194 (1964). (33) D . H. R. Barton and E. J. Tarlton, J . Chem. Soe., 3492 (1954). (34) W.A. Ayer and W. I . Taylor, ibid., 3027 (1955). (35) Cf.D.H. R. Barton and C . H. Robinson, ibid., 3045 (1954); A. J. Birch, H. Smith, and R. E. Thornton, ibid., 1339 (1957). (36) Cf.W.R. Bamford and T. 9. Stevens, ibid.. 4735 (1952). (37) M. K. Seikel and J. W. Rowe, Phytochemistry, 3, 27 (1964).
HUMBER, PINDER, AND WILLIAMS
2338
Experimental (+)-11,12-Epoxy-l1,12-dihydro-lS-nor-a-cyperone (6).-(+)15-Nor-a-cyper0ne1~1la(3.05 g) in anhydrous chloroform (15 ml) was added gradually to a freshly prepared solution of perbenzoic acid (2.11 g) in chloroform (87 ml) a t 0'. After 24 hr at 0" the solution was shaken with aqueous sodium bicarbonate and with water, dried, aiid concentrated. The residual epoxide (6) distilled a t 104-106' (0.50 mm) (2.80 g); n Z 41.5225; ~ [ a ] D +75.4' (e 3.1); 231 mp (e 13,200); v:: 1681 (conjugated ketone C=O), 1629 (conjugated C=C), and 1250 cm-l (epoxide). Anal. Calcd for C14H2002: C, 76.36; H, 9.09. Found: C, 76.12; H , 9.10. 3-Hydroxy-15-nor--,-eudesmol(7).-A cold solution of lithium aluminum hydride (0.82 g) in anhydrous ether (20 ml) wm added slowly, with agitation, to the foregoing epoxide (3.2 g) in anhydrous ether (40 ml), at 0'. After 24 hr a t 0" the mixture was decomposed in the usual way, and the product was isolated with ether. Evaporation afforded the diol 7 (3.1 g), a mixture of CI epimerides, as a syrup: bp 115-116' (0.1 mm); [ a ] D f64.8" 208 mp (e 8930); YE3510 (OH) and 1658 ( c 5.1); ?A:" crn-1 (C=C). Anal. Calcd for C14H2402: C, 75.00; H, 10.71. Found: C, 74.65; H , 10.50. (+)-15-Norcarissone (8).-The foregoing diol (2.4 g) in dry chloroform (120 ml) was shaken with freshly prepared manganese dioxide'4 (25 g) for 24 hr. The suspension was filtered and the filtrate was concentrated in vacuo, leaving (+)-lS-norcarissone (2.1 g): bp 107-108" (0.1 mm); [a]D +102.7" ( C 2.8);": : :A 241 mp (e 10,530); v:::' 3704 (OH), 1672 (conjugated ketone C=O), and 1681 cm-l (conjugated C=C). Anal. Calcd for CI4Hz2O2: C, 75.68; H , 9.91. Found: C , 75.61; H , 9.90. The 2,4-dinitrophenylhydrazonecrystallized from methanol in 390 mp (e 21,900). red needles: mp 136-137", Anal. Calcd for CjOH26N408:C, 59.70; H, 6.47; N, 13.93. Found: C, 59.76; H, 6.51; N, 14.08. (+)-15-Norcarissone Ethylene Dithioketal (9).-(+)-15Norcarissoric (1.78 g) in methanol (25 ml) was mixed with ethanedithiol (1.28 g) and rooled to 0". Freshly distilled boron trifluoride etherate (1 ml) was added; after 24 hr at 0" the methanol was removed in uacuo and the product was isolated with ether. The ethylene dithioketal distilled at 130-135" (0.2 mm) (2.1 g); it solidified easily and crystallized from petroleum ether in needles, mp 123", undepressed by admixture with a sample of the product obtained by the reaction of the ethylene dithioketal24 with methylrnagnesium iodide (see below). Condensation of ( - )-Dihydrocarvone with 1-Diethylaminobutan-3-one Methiodide.-This condensation was effected using methylsulfinyl carbanion as base,lb in a manner closely similar to that described for the (+) enantiomer.' The crude product from 21.3 g ( - )-dihydrocarvone was distilled fractionally, affording a forerun (5.3 g) of unchanged ketone, followed by a fraction, bp 132-150' (0.5 mm) (13.6 g), which crystallized partially on keeping. Trituration with petroleum ether gave the ketol (12) (3.6 g): mp 105-106' (lit. mp llO-lll°C~,,and106-107" 40); [ a ] D -56.1" ( C 2.0) [ a ] D -58°);g~s;. 3704 (OH), 1721 (C=O), 1647 '2nd 894 cm-I (C=CH2); T.IIS 1.18 (singlet, 3 H, angular CH3), 1.68 (singlet, 2 H , =CCHa), 2.18 (singlet, 1 H, OH), 2.45 (AB quartet, partly obscured, 2 H , J = 14 cps, COCHZCe), and 4.65 (broad singlet, 2 H, C=CHz); ORD in methanol ( e 0.15) [4IdoO -320', [&I304 +1130", [4]261 -4220' (sh), [&1213 -7550". (38) Melting goints and boiling points are uncorrected. Ultraviolet spectra were measured on a Unicam SP800, and infrared spectra were measured on Perkin-Elmer Infracord 137 and 237, and Unicam SP200G instruments. Nnlr spectra were determined on a Perkin-Elmer R10 spectrometer a t 60 Mcps, in deuteriochloroform solution, using tetramethylsilane as internal standard. ORD measurements were made on & Bellingham-Stanley/Bendix-Ericsson Polarmatic 62 spectropolarimeter. Glpc analyses were performed on a Pye argon chromatograph, using a 30% PEGA on 100-120 mesh Celite column, on a Wilkens "Aerograph" Model 600-C (10% Carbowax on 60-80" Chromosorh W), and on a Wilkens "Aerograph Autoprep" Model A-700 instruments. Tlc was performed on silica gel or silica gel impregnated with silver nitrate. Preparative tlc was on Merek silica gel H F 254, using 1-mm layers. Spots and bands were located with ultraviolet light. Optical rotations are in chloroform at 25'. Petroleum ether refers t o the fraction bp 60-80'. (39) D. W. Tlleobald, Tetrahedron Letters, 969 (1966) ; Tetrahedron, 92, 2869 (1966). (40) J. A. Marlhall, W. I . Fanta, and H. Roebke, J . Org. Chem., 81, 1016 (1966).
VOL.32
Anal. Calcd for Cl4H2201: C, 75.63; H, 9.97. Found: C, 75.35; H, 9.73. The noncrystalline material (ca. 10 g) on chromatography on silica gel (Davison, 200 g) gave, on elution with benzenechloroform (10: 1, followed by 3:2), a pale yellow liquid: bp 110~ ( c 2.0); ?:A: 235 mp (e 120' (0.2 mm) (2.8 g); [ a ]+77.3" 13,000); v:: 1675 (conjugated C=O), 1618 (conjugated C=C), and 893 cm-1 (C=CH2). Thin layer chromatography showed the presence of two components, the epimeric ketones 17 and 18 (see below). Elution with chloroform gave firstly an orange liquid (6.0 g), which crystallized from chloroform to give further ketol 12 (1.9 g), mp 103-104", followed by the 0-hydroxy sulfoxide 16 (0.77 g) as needles from petroleum ether: mp 115-116'; [ a ] ~ +120' ( c 2.0); A",".'l" 205 mp4I (e 1788); v:$ 3636 (free OH), 3472 (bonded OH), 1650 and 891 (C=CH2), and 1020 cm-' (S+O, H-b~nded);~X 6:20.95 (doublet, J = 4 cps, 3 H, CHCHI), 1.74 (singlet, 3 H , =CCH3), 2.60 (singlet, 3 H , CHsS), 2.77 (AB quartet, J = 12 cps, 2 H, CH2SO), 3.85 (singlet, 1 H , OH), and 4.70 (singlet, 2 H , C=CH2). Hydrolysis with dilute aqueous alkali afforded (- )-dihydrocarvone and dimethyl sulfoxide. Anal. Calcd for ClzHn02S: C, 62.58; H, 9.63; S, 13.89. Found: C , 62.15; H , 9.48; S, 14.29. Octalone 18. A.-A solution of the ketol 12 (3.0 g ) in ethanol (50 ml) containing concentrated hydrochloric acid (6 ml) was kept a t 0' for 10 days. Isolation with ether gave the octalone 18 (2.6 9): bp 110-114" (0.15 mm); [ a ]+176.0° ~ ( e 2.0) [lit. bp 85-95' (bath) (0.05 mm),@ [ a ]+185"39]; ~ ?: :A 235 mp 1678 (conjugated C=O), 1618 (conjugated (e 13,100); :::Y C=C), and 895 cm-1 (C==CH2); 8$$: 1.26 (singlet, 3 H , angular CH3), 1.72 (singlet, 3 H , C=CCH3), 4.82 (unresolved multiplet, 2 H , C=CH2), and 5.70 (singlet, 1 H , olefinic H ) ; ORD in methanol [c 0.15 (500-300 mp) and0.015 (300-250mp)], [4]500+190", [4]400+480", [@I355 +990", [+I312 -1160', [ 4 ] 2 5 0 -470'. Anal. Calcd for Cl4H2oO: C, 82.30; H , 9.87. Found: C, 82.04; H, 9.61. The semicarbazone crystallized from methanol in photochromic ~ (c 0.45, pyridine). prisms: mp 221', [ a ]+260° Anal. Calcd for C1SH23N30: C, 68.93; H , 8.87; N, 16.08. Found: C , 68.57; H, 8.72; N, 16.25. B.-The mixture of epimeric ketones 17 and 18 was converted into the semicarbazones in the usual way. The crude product (5.0 g) was digested for some minutes with ethanol (50 ml) and the suspension filtered hot. The residue (2.4 g), methanol (115 ml), water (50 ml), petroleum ether (40 ml), oxalic acid (6.5 g), and 407, aqueous formaldehyde (33 ml) were refluxed for 3 hr. Water was added and the organic layer was separated, washed with water, dried, and concentrated. The residual ketone (1.2 g) distilled a t 121-122" (0.4 mm) and was identical in all respects with octalone 18 obtained as in 8 . Ozonolysis of Octalone 18. A.-The octalone 18 (2.0 g) in dry chloroform (20 ml) was ozonized a t 0' for 15 hr. Water (100 ml) was added and the solvent was steam distilled. Then the cooled, aqueous residue was extracted several times with chloroform and the combined extracts were dried and evaporated. The residue distilled a t 140' (0.2 mm) (1.3 9). Thin layer chromatography of the partly crystalline distillate showed the presence of two components, separable by preparative thin layer chromatography, with development with benzene-ethyl acetate (8:2). The fraction with higher Rr (epoxide 20) crystallized from petroleum ether in plates (0.4 g): mp 104-105.5'; [ a ] D +4.0" (c 2.0); no ultraviolet absorption; v::: 1715 (ketone C=O), 1260 and 925 cm-l (epoxide C0);438:z1.12 (singlet, 3 H, angular methyl), 2.19 (singlet, 3 H , COCHa), and 3.09 (singlet, 1 H, oxirane ring proton); ORD in methanol ( c 0.30), [dim -390". ['$I323 -5000', [4lza2 +8600", [$JIM+5550", . . i&II +7400°. Anal. Calcd for ClaH1803: C, 70.24; H, 8.16. Found: C, 70.21; H,7.91. The fraction with higher Rf (diketone 19) crystallized from petroleum ether (bp 40-6O0)-ether in prisms: mp 56-57" (0.2 9); [ a ] D +156" ( c 2.0);" : :A 242 mp (e 14,100); Y%' 1720 (saturated ketone C=O), 1680 (conjugated ketone C = O ) , and 1620 cm-1 (conjugated C=C); 8 4 ' 2 1.26 (singlet, 3 H , angular (41) C/. H.P. Koch, J . Chem. Soc., 2892 (1950); G. Leandri, A. Mangini, and R.Passerini, ibid., 1386 (1957). (42) C/.T. Cairns, G . Eglinton, and D. T . Gibson, Spectrochim. Acta. 20, 31 (1964). (43) L. J. Bellamy. "The Infrared Bpectra of Complex Molecules," Methuen and Co. Ltd., London, 1958, p 115.
JULY1967
SYNTHESIS OF (+)-a-AND (+)-/~-EUDESMOLS
2339
1645, 847 cm-1 ( C = C ) ; 8:G;la 1.07 (singlet, 3 H, angular CH,), CH3), 2.14 (singlet, 3 H , COCH3), and 5.83 (singlet, 1 H, olefinic S H); ORD in methanol [c 0.06 (500-300 mp) and 0.03 (300-196 1.75 (singlet, 3 H, C H d X S ) , 3.30 (multiplet, 8 H, SCH2CH2S), mp)l, [+I500 +350", +380", [+la55 +1040", [+I306 -7OOo", and 5.55 (sinelet. 1 H. olefinic H): ORD in methanol-dioxane [+I256 +44000", [+I226 -9000°, [+I211 -14600", [6]196 -35000". (4: 1) (C 0.04: [+]so0 +260° [+I400'+650°, [+I219 +3900", [+I250 Anal. Calcd for Cl3HlSO2:C, 75.69; H , 8.80. Found: C, -6500°, [+I216 +25000". 75.48; H , 8.87. Anal. Calcd for C1?&S4: c, 56.97; H, 7.06; S, 35.72. B.-The octalone 18 (2.0 g) in chloroform (20 ml) and pyridine Found: C, 56.78; H, 7.39; S, 35.11. (0.8 ml, 1 mole) was ozonized for 15 hr a t 0'. Zinc dust was Reaction of Methyl Ketone 24 with Methylmagnesium Iodide. added, with ice cooling, and the mixture was shaken for 20 min, -A Grignard reagent was prepared from magnesium (0.45 g), filtered, and treated with water (100 ml). The solvents were dry ether (20 ml), and methyl iodide (3.5 g), The methyl ketone removed by steam distillation, and the residue was extracted with 24 (1.5 g) in dry ether (25 ml) was added gradually, with stirring chloroform. The extract was dried and concentrated in vacuo, a t 0"; on completion of the addition the mixture was refluxed to yield a pale yellow liquid, bp 140' (0.3 mm) (1.6 g), which for 30 min on the steam bath, cooled, and decomposed with satusolidified readily and crystallized from petroleum ether (bp rated aqueous ammonium chloride. The ether layer was sepa40-6O0)-ether in prisms, mp 55-56', identical with the diketone rated, washed with water, dried, and concentrated, leaving the 19 described above. tertiary alcohol 9, which crystallized from petroleum ether in Epimerization of Diketone 19.-The diketone (100 mg) in ethaneedles (0.95 g): mp 123-123.5'; [ a ] f93.6" ~ (c 2.0); A.: nol (2 ml) containing concentrated hydrochloric acid (0.2 ml) was kept for 24 hr, then poured into water. Ether extraction 244 mp ( B 800);44 v::? 3620 (free OH), 3500 (bonded OH), and 1645 cm-l (C=C); 6:;; 1.07 (singlet, 3 H, angular CH3), 1.14 afforded, after evaporation, diketone 21 as needles from petro[singlet, 6 H, C(CH3),], 3.28 (multiplet, 4 H, SCHZCHZS),and leum ether (42 mg): mp 107-107.5"; [ ~ J D$83" (e 2.0);": :A 5.48 (singlet, 1 H, olefinic H ) ; ORD in methanol (c 0.065), 238 mp (e 16,080); ~2: 1720 (saturated ketone C=O), 1680 (conjugated ketone C=O), and 1620 cm-* (conjugated C=C); ['$]so0 +350", [6]~io +400", [+I270 $3970, [@]z49-50000'. Anal. Calcd for CI6Hz60S2:c, 64.40; H , 8.78; S, 21.45. ;::6 1.26 (singlet, 3 H, angular CHs), 2.12 (singlet, 3 H , COCHs), Found: C, 64.20; H , 8.70; S, 21.51. and 5.62 (singlet, 1 H , olefinic H ) ; ORD in methanol [c 0.074 15-Nor-~-eudesmol(10). A.S-To a stirred solution of the (500-270 mp) and 0.024 (270-195 mp)], [+I500 +190", [+]a00 ethylene dithioketal 9 (500 mg) in dry ether (20 ml) and liquid +310", [+I381 -290', [6]300+2800", [6]z70 +14300°, [+]zag ammonia (50 ml) was added sodium metal (500 mg). The mix+32000", [+I218 -39000", [+I195 -31400". ture was stirred for 10 min and then treated with sufficient ethanol Anal. Calcd for C13H1802: C, 75.69; H, 8.80. Found: C, to dispel the blue color. The ammonia was allowed to evaporate, 75.39; H , 8.96. and water and ether were added. The ether layer was separated, Ozonolysis of Keto1 12.-The keto1 12 (3.0 g) in dry chloroform washed with water, dried, and concentrated in uacuo. The (20 ml) was ozonized for 18 hr at 0". Water (100 ml) was added residual 15-nor-~-eudesmol(10) distilled a t 88-90' (0.15 mm) and the solvent was steam distilled. Then the cooled residue (395 mg) (single peak on gas chromatography): [ a ]+47.8" ~ was extracted with chloroform, and the dried extract was con( c 2.0); 205 mp ( e 6040); 3571 (OH), 1661, 807 cm-1 centrated in vacuo, leaving the diketo alcohol 23 (2.9 g), which crystallized from petroleum ether-benzene (1 :1) in needles: ( R I R ~ C ~ H R6 ~ :)z ; 1.05 (singlet, 3 H, angular CHB), 1.20 ~ (e 2.0); 3610 (free OH), 3470 [singlet, 6 H, C(CH,),], and 5.34 (broad, 1 H, =CH); ORD in mp 142-143'; [ a ]-42" (bonded OH), and 1715 cm-1 (saturated ketone C d ) ; 6 ~ ~ ~ - c c l rmethanol (c 0.121, [+Is~o + l l O " , [dl400+180", [+I300 +480", 1.20 (singlet, 3 H, angular CH3), 2.12 (singlet, 3 H , COCHa), [@]250+1090", [@]z19+1900". Anal. Calcd for C14H240: C, 80.71; H , 11.61; 0, 7.68. and 2.20 (singlet, 1 H, OH); ORD in methanol (cO.O65), [+I500 Found: C, 80.36; H , 11.66; 0, 8.09. +350", [+]400 +610°, [+]m +3970", [+I249 -5000", [+1218 B.-W-2 Raney nickel was partially poisoned by refluxing in +25000". 90y0 aqueous ethanol containing pyrrolidine ( 1 ml per 3 g of Anal. Calcd for C1&003: C, 69.61; H , 8.99. Found: C, nickel) for 5 hr.46 The catalyst was washed well by decantation 69.63; H, 8.97. with 907, ethanol. The ethylene dithioketal 9 (400 mg) was Dehydration of Diketo Alcohol 23.-The above diketo alcohol refluxed with the poisoned catalyst (ca. 6 g) for 16 hr under (5.8 g) in ethanol (80 ml), containing concentrated hydrochloric acid (10 ml), was kept at 0" for 7 days. Isolation with ether nitrogen. Removal of the catalyst and concentration of the filtrate gave 15-nor-y-eudesmol, bp 90' (0.16 mm) (251 mg), gave a semicrystalline mass (4.9 g) of the diketone 21, which identical with that obtained as described under A . separated from petroleum ether in needles, mp 106-107", identiHydroboration-Oxidation of 15-Nor-y-eudesmol .-The apparacal with the product described above (mixture melting point tus described by Brown and Zweifel16 for the external generation comparison). Reaction of Diketone 21 with Ethanedithio1.-Diketone 21 (4.8 of diborane was used. Boron trifluoride etherate (15 ml) in dry g) and ethanedithiol (1.95 ml) were dissolved in methanol (60 diglyme (10 ml) was added dropwise to a stirred solution of soml) and treated a t 0" with boron trifluoride etherate (3.3 ml). dium borohydride (1.14 g) in diglyme (20 mi). The diborane formed was swept by a slow nitrogen stream through a solution After 24 hr at 0" water (60 ml) was added and the product was isolated by chloroform extraction. Evaporation of the washed of 15-nor-y-eudesmol (272 mg) in dry tetrahydrofuran (10 ml), (NaHC03) and dried extracts in vacuo left a viscous, yellow liquid a t 0". Toward the end the diborane generator was heated to (3.7 g), which was chromatographed on alumina (Woelm neutral, 70-80'. The tetrahydrofuran solution was treated with water activity 111, 80 9). Elution with petroleum ether-benzene (4: 1) ( 5 ml), 4 N aqueous sodium hydroxide (10 ml), and 30% hydroafforded, aft,er solvent evaporation, a waxy solid (2.0 g), which gen peroxide (10 ml), with shaking and ice cooling, and kept crystallized from petroleum ether in prisms of the ethylene dithioovernight a t room temperature. Water and ether were added, f50.5' (c 2.0); ?A: 245 mp ketal 24: m 85-86.5"; [.ID and the aqueous layer was saturated with salt. The product, a mixture of epimeric diols 26, was isolated by several ether ex(e "2;: 1715 (saturated ketone C = O ) , 1645 cm-'(C=C); tractions: evaporation afforded a viscous liquid (301 mg), 8::; 1.08 (singlet, 3 H, angular CH3),2.07 (singlet, 3 H, COCHS), 3.26 (multiplet, 4 H , SCHzCHB), and 5.47 (singlet, 1 H , olewhich showed no C=C absorption in the infrared region. The product was dissolved in acetone (10 ml) and titrated with finic H); ORD in methanol (c 0.04), [+]E~o$180', [+I400 +220", Jones reagentz6 at 0" until an orange color persisted. After [@I302 -990°, [+I269 -3900", [+I249 -2300", [$]m -31000". keeping for 20 min water was added and the product was isolated Anal. Calcd for CI;H2zOS2: C, 63.81; H, 7.85; S, 22.67. Found: C, 63.57; H , 7.94; S, 23.03. with ether. Evaporation of the solvent left a yellow liquid (280 Further elution of the chromatogram with petroleum ether-benmg), a mixture of the expected czs- and trans-I-decalones. The Bene ( 1 : 1) gave a gum, which was triturated with ether and kept crude product, in benzene, was poured onto a column of Woelm basic alumina (activity I , 10 g); after 24 hr, elution with benzenea t 0" for several hours. A crystalline solid (0.4 g) separated; it chloroform ( 1 : 1) was effected, affording, after evaporation, 78was collected, washed with ether, and purified by preparative thin layer chromatography. Elution with petroleum ether (bp ( 1-hydroxy - 1-methylethyl) -106-methyl-trans-4-decalone (27), 40-60") and crystallization from petroleum ether-benzene (1 : 1) which crystallized from petroleum ether in prisms: mp 119-120' gave the bis(ethy1ene dithioketal) 25 (0.4 g) as prisms: mp 170(225 mg); [ a ] D $7.1" ( c 2.0); v$:'~ 3600 (free OH), 3450 248 mp ( e 1700);44 ~f::' 170.5'; [alD +93.7" ( c 2.0);?: :A
v::
v::?
(44)
Cf.D. A. Lightner, C. Djerassi, K. Takeda, K. Kuriyama, and T.
Komeno, Tetrahedron, 21, 1581 (1965).
(45) Compare W. Oroshnik, G. Karmas, and A . D. Mehane. J . A m . Chem. Sac., 74, 295 (1952). (46) H. C . Brown and G. Zweifel, ibid., 83. 2544 (1961).
2340
HUMBER,PINDER, AND WILLIAMS
VOL.
32
&:sa [+]tea -4550" sh,
(bonded OH), and 1705 cm-' (saturated ketone C d ) ; 0.78 (singlet, 3 H, angular CHJ), 1.20 [singlet, 6 H, C(CH&], and 1.63 (singlet, 1 H, OH); ORD in methanol (c 0.09), [+]m +50", [+]a00 -50°, [+I301 -2150', [+I264 +3450', [+]no +1500". This product was identical in all respects with the ketone obt i n e d by ozonolysis of commercial eudesmol (see below). Anal. Calcd for C1aHz402: C , 75.00; H, 10.71. Found: C, 75.06; H, 10.60. Ozonolysis of Commercial Eudesmol.Qa~2*--Eudesmol ( 1 .O g) in dry chloroform (15 ml) was ozonized for 6 hr a t 0 ' . Water was added and the solvent was steam distilled. The aqueous residue was cooled and extracted several times with chloroform. Evaporation of the combined, dried extracts afforded a liquid residue (1.O g), thin layer chromatography of which showed the presence of several components. The product, in benzene, was chromatographed on Woelm neutral alumina (activity 111, 25 9). Elution with benzene yielded the ketol 27 (95mg), which solidified, and crystallized from petroleum ether in prisms, mp 118119' (lit.2*mp 119-120"),identical in all respects with the synthetic material described above. (+)-pBudesmol(2) .-Dimethyl sulfoxide (4ml) was added to powdered sodium hydride (160 mg) and the whole was stirred a t 75-80' under dry nitrogen for 45 min. The clear solution was cooled in ice-water, and methyltriphenylphosphonium bromide (2.4g) in dimethyl sulfoxide (5ml) was added. The yellow ylid solution was stirred a t ambient temperature for 10 min; the ketol 27 (150mg) in dimethyl sulfoxide (4ml) was then added, and the mixture was stirred a t 50-60' for 30 hr under nitrogen.47 The cooled system was decomposed with ice-water, and the product was isolated with ether. The extracts were well washed with water, dried, and concentrated. The residue was a crystalline mass (356 mg) containing an appreciable amount of triphenylphosphine oxide. It was resolved by preparative thin layer chromatography, with elution by benzene-ethyl acetate (9:1). The more mobile fraction was a white, waxy solid (113 mg), identical in all respects (infrared spectrum, tlc mobility, glpc retention time, nmr spectrum) with natural ( +)-p-eudesmol.m The synthetic. product was further purified by crystallization from petroleum ether (bp 40-60"),from which it separated in spongy needles, mp 79-80", or by vacuum sublimation, as needles: mp 79-80"; [ a ] D +56.6" (c 2.0) (lit.308mp 80-81', [ a ] +58.0"); ~ ;6 0.72 (singlet, 3 H , angular CHI), 1.18 singlet, 6 H, C(CHE),], 4.47 and 4.72 (singlets, 2 H , C=CH,); 3618 (free OH), 3480 (bonded OH, broad), 1650 and 889 (C=CH2), 1380 and 1370 cm-' [C(CH&] [lit.?b *':;:6 0.69 (angular CHI), 4.42 and 4.77 (=CH*); lit.308 0.7 (angular CHI), 1.2 [C(CH&], 4.47 and 4.73 ppm (C=CH*)], ( -)-trans-Dihydrocarissone (29) .-( +)-Carissone (28) was isolated from tubers of Carissa Zan~eoluta,~* as described by Mohr. Schindler, and Reich~tein.'~The ketone (472mg) in dry ether (10 ml) was added to a stirred solution of lithium metal (100mg) in liquid ammonia (100ml) . After 40 min, solid ammonium chloride was added and the ammonia was allowed to evaporate. Addition of water and ether extraction gave, after evaporation, a solid, mp 72-82' (451mg), thin layer chromatography of which showed the presence of one major and two minor components. The crude product (448 mg) in purified acetone (10 ml) was titrated a t 0" with Jones reagent.2e Dilution nrith ice-water and ether extraction afforded a crystalline solid, mp 107-108" (415mg), showing a single spot on tlc. A benzene solution of the product was filtered through alumina and evaporated, and the residue was Crystallized from petroleum ether, from which the product separated in needles: mp 109-110"; ID -11.7' ( c 2.0);:!:v 3615, 1715 cm-1; 6;:; 6.91 (singlet, 3 H, angular CHI), 1.02 and 1.10 (doublet, J = 5 cps, 3 H, C-4-CHs), 1.18 [singlet, 6 H , (CH&C], and 2.42 (singlet, 1 H , OH); ORD in methanol ( c 0.10),[+I400 -50", [+I303 +2500",
f&
(47) Compare R. Greenwald, M. Chaykovsky, and E. J. Corey. J . Org. Chem., 98, 1128 (1963). (48) We wish t o thank Professor T. Reichstein. Basel, for generous gifts of (+)-carissone and Cartsaa lancedata tubers. (49) K. Mohr, 0. Schindler, and T. Reichstein, Helu. C h i n . Acta, 37, 462 (1954).
[+]ma -5250'. Subsequent experimentation showed that the crude reduction product (238 mg) could be satisfactorily purified by equilibration with concentrated hydrochloric acid (0.2ml) in ethanol (5ml) for 12 hr a t 0'. Anal. Calcd for C15H1602: C, 75.58; H, 11.00. Found: C, 75.61; H, 10.86. trans-Dihydrocarissone Tosybydrazone (31).-A solution of the foregoing ketone (100 mg), toluene-p-sulfonylhydrazine (90 me), and concentrated hydrochloric acid (1 drop) in ethanol (5 ml) was refluxed for 1 hr. Removal of the solvent in vacuo gave the hydrazone which crystallized from benzene in needles: mp 143-144' dec (143 mg), [ a ]4-37.7" ~ ( c 2.0). Anal. Calcd for C22H34N203S: N, 6.89; S, 7.87. Found: N, 6.81; S, 7.97. ( +)-a-Eudesmol ( I ) .-The above tosylhydrazone (230 mg) in ethylene glycol (10 ml) containing dissolved sodium metal (250 mg) was heated under nitrogen a t 170-190" (bath) for 1 hr, then poured into water.3e The product, isolated with ether, was a pale yellow solid (151 mg); it was sublimed twice a t 16 mm (bath temperature 90') to give needles of substantially pure ( +)-a-eudesmol (51 mg) . Thin layer and gas chromatography of the product, and its nmr spectrum, indicated that it contained a trace of impurity. Further purification was effected by a combination of preparative thin layer (on silica gel impregnated with silver nitrate50) and gas chromatography, leading to pure (+)-a-eudesmol (45 mg): mp 74-75'; [ a ] +28.5' ~ (c 1.2) (lit.?a mp 75", [ a ] f28.6"); ~ 0.79 (singlet, 3 H , angular CHI), 1.22 [singlet, 6 H, (CH&C], 1.47 (singlet, 1 H,OH), 1.63 (broad singlet, 3 H, =CCH3), and 5.45 (unresolved multiplet, 1 H, =CH) (lit.?b 6:gp 0.76 and 5.27); ORD in methanol ( c 0.05), [+I400 +zoo, [+]so0 +15", [+lzso -go", [+1216 -2600". This product was identical in every respect (tlc, glpc, and infrared comparison) with an authentic specimen of (+)-aeudesmol isolated from a sample of commercial eudesmol, containing a and p isomers only, as described by Seikel and R ~ w e . ~ ' Alternatively, separation was effected more rapidly and efficiently by preparative thin layer chromatography on silica gel impregnated with silver nitrate!* For this purpose the silica gel plates were sprayed with 13% aqueous silver nitrate, then dried a t 120' for 30 min, cooled, and used immediately. They were streaked manually with 25-30 mg of commercial eudesmol per plate (20 X 20 cm, layer 1 mm thick), and developed with benzeneethyl acetate (8:2). Two zones were observed; the upper one, on removal, extraction with chloroform, concentration, and sub~ limation, afforded (+)-a-eudesmol: mp 74-75', [ a ]+29.0". Anal. Calcd for C15H260: C, 81.02; H, 11.79. Found: C, 80.46; H, 11.97.
Registry No.-1, 473-16-5; 2, 4895-24-3; 6, 1306510-6; 7, 13065-11-7; 8, 13085-21-7; 8 2,4dinitrophenylhydrazone, 13094-93-4; 9, 4895-28-7 ; 10, 4893-29-8 ; 12, 4895-25-41 16, 13065-14-0; 17, 13005-23-9; 18, 520862-8; 18 semicarbazone, 13065-16-2; 19, 13065-17-3 ; 20, 13065-18-4; 2 1,4895-27-6 ; 23,4895-26-5; 24, 5003-95-2 ; 25, 13094-94-5; 27, 4895-32-3; 29, 13065-22-0; 31, 13085-25-1. Acknowledgments.-We are grateful to Imperial Chemical Industries (Fibers) Ltd. for the award of a research studentship (to D. C. H.) and to the Department of Scientific and Industrial Research (Great Britain) for a maintenance grant (to R. A. W.). We thank the Government Grants Committee of the Royal Society (London) for a grant for the purchase of equipment. Dr. A. Blummn (Plaimar Ltd., Perth, W. Australia) kindly provided samples of commercial eudesmol. We acknowledge the helpful comments of Dr. R. P. Houghton. (50) CJ'., infer alia, A. S. Gupta and S. Dev. J . Chromafog.,19, 189 (1963); E. von Rudloff,Chem. Ind. (London), 2126 (1964).
